Study finds AI chatbots give wrong election information and pick candidates
11/03/2026 — 11/03, 04:17·1 sources (all)·1 reports (all)
Key takeaways
- NPR reports that as US midterms approach, more voters are asking AI chatbots about elections, and researchers found several popular products…
Background & analysis
According to a report by NPR on November 3, 2026, voters in the United States are increasingly turning to AI chatbots for information ahead of the midterm elections. Citing a Pew Research Center survey from earlier this year, the report says that about half of American adults say they have used an AI-powered chatbot and that about 42% of users say they use them to search for information.
Researchers found that many of the most popular chatbots may provide misleading information about candidates and about polling site opening times, and that some have even provided lists of candidates to vote for. Experts expressed concern that AI-generated misinformation could shape voters' decisions. The nonprofit organization IFPG is testing seven chatbots, including Google AI Mode and AI Overviews, Gemini, ChatGPT, Claude, Copilot and Perplexity. The available reporting does not yet detail the specific test results, and the situation currently stands at the testing stage.
AI-generated from 1 reports · updated 3 days ago
- Heat index
- 25
- All sources
- 1
- All reports
- 1
- First seen
- 3 days ago
Reports on this story headlines open the original
NPR reports that as the midterms approach, growing numbers of US voters are asking AI chatbots for election information, yet researchers found popular models giving wrong candidate details and polling-place opening times, with some even listing candidates to vote for. The nonprofit Innovating for the Public Good is testing seven major chatbots, including Google AI Mode and AI Overviews, Gemini, ChatGPT, Claude, Copilot and Perplexity.
NPR · 科技AI score 54
Other stories people are talking about
- Heat index 840OpenAI ships Images 2.6 with GIF and SVG output15 sources
- Heat index 675OpenAI says its cyber model pipeline confirmed 1,724 Linux kernel flaws and generated 948 privilege-escalation exploits11 sources
- Heat index 642RisingOpenAI GPT-5.6 Brings Major Agentic and Computer Use Upgrades13 sources
- Heat index 633RisingCat Wu, product lead for Claude Code, joins OpenAI9 sources
- Heat index 630NewSurgeOpenAI releases GPT-6 Cyber cybersecurity model7 sources
How is heat calculated?About the methodHide
Heat counts how many independent sources covered a story in the last 48 hours: one source counts once no matter how many posts it published, decaying with a 24-hour half-life. What ranks first is what many people are talking about.
This page aggregates public feeds. Headlines and summaries are machine-organized and remain the property of the original authors; verify important facts at the source. If you believe a headline or summary infringes your rights, email the contact address in the footer with the page URL and basis of your claim, and we will remove or replace it after review.
- Surge
- Discussion rising fast
- New
- First report within 6 hours
- Rising
- Still gathering discussion